e Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
Microsoft PowerPoint - Kirkpatrick [Read-Only] Is Solid State the Future of
Lighting?
Is Solid State the Future of
Lighting?
Dr. Doug Kirkpatrick
DARPA / ATO
November 2003
Dr. Doug Kirkpatrick
DARPA / ATO
November 2003
Approved for Public Release, Distribution Unlimited Approved for Public Release, Distribution Unlimited
Outline
Outline
Review of history of innovation in lighting
Solid state is not the first new lighting technology
History is a great teacher of the wrong things to do
Conventional technology and the current market
Where are we now
The other guys arent standing still
Near term targets of opportunity
Key technology steps
Approved for Public Release, Distribution Unlimited Approved for Public Release, Distribution Unlimited
Some History
Some History
Technology assaults on the lighting market are
not new
Edisons incandescent (c. 1900)
Fluorescent (c. 1930)
HID (c. 1960)
CMH (c. 1995)
Net present value of total investment is huge
100 years of manpower and capital
Key focus has been manufacturing technologies
Focal point of automation processes even before electronics
No-touch
How else does an incandescent lamp cost <7.5¢ to make?
Image provide by USDOI Approved for Public Release, Distribution Unlimited
Metrics of
The (Ever) Better Light Source
Metrics of
The (Ever) Better Light Source
Cost
First cost
Operating cost
Life cycle cost
inclusive of
Efficiency
Socket compatibility
Fixture integrability
Maintenance
Lifetime
Performance
Color
Color quality
Output
Output intensity
Optical extent
Turn-on time
Re-strike time
Operating environment
Capabilities
Characteristics Approved for Public Release, Distribution Unlimited
Performance and Cost are Drivers
Performance and Cost are Drivers
Performance and cost determine
Which market segments a technology will penetrate
How fast and how far
Performance characteristics
Necessary but not sufficient
Gating function on participation in a given segment
Cost characteristics
Determine competitive position
Drive rate and depth of segment penetration
Different market segments apply very different
weightings Approved for Public Release, Distribution Unlimited
Total Cost of Ownership
Total Cost of Ownership
Nordhaus observation
10x drop in cost per 50 years
Decrease is:
¾: increase in luminous efficacy
¼: decrease in cost of fuel
Rollover in TCO for incan-
descent lamps presaged large
shift to fluorescent lamps.
Fluorescent lamps had
substantial presence for 20y
prior.
Data for Fire and Incandescence modified from W.D. Nordhaus,
in T.F. Breshnahan and R.J. Gordon, Eds., The Economics of
New Goods (U of Chicago Press, 1997) pp. 29-70.
Data for SSL-LEDs taken from 2002 U.S. SSL Roadmap.
1.E-02
1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1800
1850
1900
1950
2000
2050
2100
Year
O
w
n
er
sh
i
p
C
o
st
(
$
/
M
L
m
-
h
r
)
10
-2
10
-1
10
0
10
1
10
2
10
3
10
4
10
5
Fire
Incandescent
HID
Fluorescent
SSL Approved for Public Release, Distribution Unlimited
History of Light Source Efficacy
History of Light Source Efficacy
0.01
0.10
1.00
10.00
100.00
1000.00
1850
1900
1950
2000
2050
Year
E
f
f
i
cacy (
L
P
W
)
?
Fire
Incandescent
HID
Fluorescent /
SSL
DATA DERIVED FROM:
Nordhaus, William D. in Breshnahan, Timothy F.//Gordon, Robert J., ed, The Economics of New Goods, pp 29-
70, The University of Chicago Press, 1997
SSL-LED Roadmap 2002 Approved for Public Release, Distribution Unlimited
Rapidly Improving LED Performance
Rapidly Improving LED Performance
0.01
0.10
1.00
10.00
100.00
1000.00
1850
1900
1950
2000
2050
Year
E
f
f
i
cacy (
L
P
W
)
?
Fire
Incandescent
HID
Fluorescent /
White LED
Red LED
Blue LED
Green LED
SUNLIGHT
DATA DERIVED FROM:
Nordhaus, William D. in Breshnahan, Timothy F.//Gordon, Robert J., ed, The Economics of New Goods,
pp 29-70, The University of Chicago Press, 1997
SSL-LED Roadmap 2002
Bergh, A//Craford, M.G.//Duggal, A.//Haitz, R. Physics Today 54, 42-47 (DEC 2001) Approved for Public Release, Distribution Unlimited
Yet Incandescents Retain
Significant Market Share
Yet Incandescents Retain
Significant Market Share
TOTAL ENERGY CONSUMPTION FOR LIGHTING (2001)
By Building Sector
By Light Source
Fluorescent
41%
HID
17%
Incandescent
42%
Fluorescent
41%
HID
17%
Incandescent
42%
Residential
Buildings
27%
Commercial
Buildings
51%
Industrial
14%
Outdoor
Stationary
8%
Residential
Buildings
27%
Commercial
Buildings
51%
Industrial
14%
Outdoor
Stationary
8% Approved for Public Release, Distribution Unlimited
Incandescent Lamps Account for More
Energy Use than Any Other Type
Incandescent Lamps Account for More
Energy Use than Any Other Type
0
5,000
10,000
15,000
20,000
25,000
Commercial
Residential
Industrial
Outdoor
Stationary
Teralumen-hours/yr
quads
0.0
1.0
2.0
3.0
4.0
5.0
Commercial
Residential
Industrial
Outdoor
Stationary
Incandescent
Fluorescent
High Intensity
Discharge
Total: 8.2 quads
quads
0.0
1.0
2.0
3.0
4.0
5.0
Commercial
Residential
Industrial
Outdoor
Stationary
Incandescent
Fluorescent
High Intensity
Discharge
Total: 8.2 quads
0.0
1.0
2.0
3.0
4.0
5.0
Commercial
Residential
Industrial
Outdoor
Stationary
Incandescent
Fluorescent
High Intensity
Discharge
Total: 8.2 quads Approved for Public Release, Distribution Unlimited
Whence CFLs?
Whence CFLs?
Nationwide home improvement store example:
13 sales fronts for light bulbs
8.5 sales fronts for incandescents
4 sales fronts for fluorescents
0.5 sales fronts for CFLs
Of the 8.5 sales fronts for incandescent lamps
3.5 fronts of candle-look-alike incandescents (no CFL alternative)
3 fronts of beam-formed incandescents (no CFL alternative)
$5-10/kilolumen
2000 hr lifetime
Delivered efficacies max 15 LPW
2 fronts of standard A-line incandescents
24¢ vs. $5 for CFL
1000 hrs life vs. 7000 hrs life for CFL
Total Cost of Ownership favors
CFL if Energy Cost > 1¢/kW-hr
yet CFLs still have only minimal impact on energy use WHY? Approved for Public Release, Distribution Unlimited
The Importance of
Color and Color Quality
The Importance of
Color and Color Quality
Color
Correlated Color Temperature (CCT)
Incandescent ~3100K
~same as warm white fluorescent
CCT(CFL)
CCT(incandescent)
Daylight ~5000-7000K (D
65
)
Color quality: Color Rendering Index (CRI)
Measure the spectral power at several discrete wavelengths
Ra(8) uses eight wavelengths (de facto standard)
Ra(14) uses fourteen wavelengths (seldom used)
Compare to spectral power of equivalent ideal grey body
Incandescent lamps feature CRI=100 by definition
CRI(CFL) ~ 80
CRI differences of 10 or more are observable Approved for Public Release, Distribution Unlimited
Gaming CRI
Gaming CRI
Eight wavelength measurement rooted in
antiquity
Predates calculators (and computers)
Eased calculation burden
Eased measurement burden
Wavelengths for spectral power measurement
fixed by standard
Sparse evaluation of spectral power density
distribution function drives gaming
Phosphor performances optimized to Ra(8) standards
HID fill chemistries optimized to Ra(8) standards Approved for Public Release, Distribution Unlimited
Evolving Color Metrics
Evolving Color Metrics
Computational limitations for evaluating light
source color quality are no longer relevant
Alternative measure: % deviation from daylight
Measured output spectrum aggregated into 10nm bins
CCT used to compute CIE equivalent daylight spectrum
Find ±% displacement necessary to form envelope that
contains all of the binned output levels of the light source
420-650 nm
Three peak bins allowed to exceed upper envelope
Null-light bins not allowed
% displacement necessary denoted by mm%Dxx
10%D65 contains all bin levels within ±10% of the CIE D65
spectrum, with maximum three peaks that exceed the +10%
envelope. Approved for Public Release, Distribution Unlimited
High Efficiency and High Color
High Efficiency and High Color
Wavelength / nm
400
450
500
550
600
650
700
S
pectral
Power /
mW-nm
-1

0
100
200
300
400
500
600
Wavelength / nm
400
450
500
550
600
650
700
Power /
mW
-n
m
-1

0
20
40
60
80
100
120
140
160
180
200
Compact HID lamp currently in development in DARPA
HEDLight program.
Current efficacy exceeds CFLs (>60 LPW)
High brightness
High efficiency optical coupling to fibers or beam-forming
optics Approved for Public Release, Distribution Unlimited
and OOPS, the Lumen May be Wrong,
Too
and OOPS, the Lumen May be Wrong,
Too
Derivation of the Lumen
Average response of a small
population of young, northern
European viewers